# Importing libraries
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns

# Loading excel file into dataframe(df)
df = pd.read_csv(r'C:\Users\Lenovo\OneDrive\Y2S3\Introduction to Data Science\Assignment\Dataset\owid-covid-data.csv')

# Checking the datatypes for each column
df.dtypes

iso_code                                  object
continent                                 object
location                                  object
date                                      object
total_cases                              float64
new_cases                                float64
new_cases_smoothed                       float64
total_deaths                             float64
new_deaths                               float64
new_deaths_smoothed                      float64
total_cases_per_million                  float64
new_cases_per_million                    float64
new_cases_smoothed_per_million           float64
total_deaths_per_million                 float64
new_deaths_per_million                   float64
new_deaths_smoothed_per_million          float64
reproduction_rate                        float64
icu_patients                             float64
icu_patients_per_million                 float64
hosp_patients                            float64
hosp_patients_per_million                float64
weekly_icu_admissions                    float64
weekly_icu_admissions_per_million        float64
weekly_hosp_admissions                   float64
weekly_hosp_admissions_per_million       float64
new_tests                                float64
total_tests                              float64
total_tests_per_thousand                 float64
new_tests_per_thousand                   float64
new_tests_smoothed                       float64
new_tests_smoothed_per_thousand          float64
positive_rate                            float64
tests_per_case                           float64
tests_units                               object
total_vaccinations                       float64
people_vaccinated                        float64
people_fully_vaccinated                  float64
new_vaccinations                         float64
new_vaccinations_smoothed                float64
total_vaccinations_per_hundred           float64
people_vaccinated_per_hundred            float64
people_fully_vaccinated_per_hundred      float64
new_vaccinations_smoothed_per_million    float64
stringency_index                         float64
population                               float64
population_density                       float64
median_age                               float64
aged_65_older                            float64
aged_70_older                            float64
gdp_per_capita                           float64
extreme_poverty                          float64
cardiovasc_death_rate                    float64
diabetes_prevalence                      float64
female_smokers                           float64
male_smokers                             float64
handwashing_facilities                   float64
hospital_beds_per_thousand               float64
life_expectancy                          float64
human_development_index                  float64
excess_mortality                         float64
dtype: object

# Checking the size/dimension of the dataset
df.shape

(96645, 60)

# Displaying the summary of each column
df.describe()

	total_cases	new_cases	new_cases_smoothed	total_deaths	new_deaths	new_deaths_smoothed	total_cases_per_million	new_cases_per_million	new_cases_smoothed_per_million	total_deaths_per_million	...	extreme_poverty	cardiovasc_death_rate	diabetes_prevalence	female_smokers	male_smokers	handwashing_facilities	hospital_beds_per_thousand	life_expectancy	human_development_index	excess_mortality
count	9.338700e+04	93384.000000	92369.000000	8.332700e+04	83483.000000	92369.000000	92889.000000	92886.000000	91876.000000	82842.000000	...	58663.000000	86900.000000	89064.000000	68010.000000	67015.000000	43663.000000	79206.000000	91773.000000	87077.000000	3407.000000
mean	1.013389e+06	6048.155294	6078.983893	2.770536e+04	147.156271	132.103750	12466.525473	76.148775	76.458143	278.009890	...	13.392748	258.343287	7.915468	10.555165	32.683993	50.848538	3.028642	73.232493	0.727521	18.243505
std	6.966606e+06	37955.391475	37720.752239	1.651512e+05	805.017702	746.455626	23160.290191	199.635466	158.986921	506.105822	...	19.955210	118.975576	4.129000	10.448992	13.481539	31.759905	2.459618	7.558983	0.150369	36.225191
min	1.000000e+00	-74347.000000	-6223.000000	1.000000e+00	-1918.000000	-232.143000	0.001000	-2153.437000	-276.825000	0.001000	...	0.100000	79.370000	0.990000	0.100000	7.700000	1.188000	0.100000	53.280000	0.394000	-95.590000
25%	1.187500e+03	2.000000	7.429000	5.200000e+01	0.000000	0.000000	248.584000	0.207000	1.270000	7.863000	...	0.600000	167.295000	5.310000	1.900000	21.600000	19.351000	1.300000	67.920000	0.602000	0.295000
50%	1.297300e+04	73.000000	90.857000	3.640000e+02	2.000000	1.286000	1688.617000	8.219500	10.883500	49.320000	...	2.200000	242.648000	7.110000	6.300000	31.400000	49.839000	2.400000	74.620000	0.748000	7.390000
75%	1.371405e+05	801.000000	841.286000	3.573000e+03	18.000000	14.286000	12800.104000	70.419000	78.928000	301.538000	...	21.200000	329.635000	10.080000	19.300000	41.100000	83.241000	3.861000	78.740000	0.848000	23.795000
max	1.777647e+08	906008.000000	826389.571000	3.849865e+06	18050.000000	14737.000000	179149.680000	18293.675000	4083.500000	5760.460000	...	77.600000	724.417000	30.530000	44.000000	78.100000	100.000000	13.800000	86.750000	0.957000	409.690000

8 rows × 55 columns

# Displaying the top records of the dataset
df.head()

	iso_code	continent	location	date	total_cases	new_cases	new_cases_smoothed	total_deaths	new_deaths	new_deaths_smoothed	...	extreme_poverty	cardiovasc_death_rate	diabetes_prevalence	female_smokers	male_smokers	handwashing_facilities	hospital_beds_per_thousand	life_expectancy	human_development_index	excess_mortality
0	AFG	Asia	Afghanistan	2020-02-24	1.0	1.0	NaN	NaN	NaN	NaN	...	NaN	597.029	9.59	NaN	NaN	37.746	0.5	64.83	0.511	NaN
1	AFG	Asia	Afghanistan	2020-02-25	1.0	0.0	NaN	NaN	NaN	NaN	...	NaN	597.029	9.59	NaN	NaN	37.746	0.5	64.83	0.511	NaN
2	AFG	Asia	Afghanistan	2020-02-26	1.0	0.0	NaN	NaN	NaN	NaN	...	NaN	597.029	9.59	NaN	NaN	37.746	0.5	64.83	0.511	NaN
3	AFG	Asia	Afghanistan	2020-02-27	1.0	0.0	NaN	NaN	NaN	NaN	...	NaN	597.029	9.59	NaN	NaN	37.746	0.5	64.83	0.511	NaN
4	AFG	Asia	Afghanistan	2020-02-28	1.0	0.0	NaN	NaN	NaN	NaN	...	NaN	597.029	9.59	NaN	NaN	37.746	0.5	64.83	0.511	NaN

5 rows × 60 columns

# New features
df['CFR'] = (df['total_cases'] / df['total_deaths'])*100
df['CFR_per_million'] = (df['total_cases_per_million'] / df['total_deaths_per_million'])*100
df['impact'] = (df['total_cases'] * 0.1 + df['hosp_patients'] * 0.2 + df['icu_patients'] * 0.3 + df['total_deaths'] * 0.4)
df['impact_per_million'] = (df['total_cases_per_million'] * 0.1 + df['hosp_patients_per_million'] * 0.2 + df['icu_patients_per_million'] * 0.3 + df['total_deaths_per_million'] * 0.4)

# Update graph font size
plt.rcParams.update({'font.size': 22})

# Function to plot a heat map
def heat_map(dataframe):
    plt.figure(figsize = (20,25))
    corr = dataframe.corr()
    sns.heatmap(corr[['total_vaccinations', 'people_vaccinated', 'people_fully_vaccinated', 'new_vaccinations', 'new_vaccinations_smoothed', 'total_vaccinations_per_hundred', 'people_vaccinated_per_hundred', 'people_fully_vaccinated_per_hundred', 'new_vaccinations_smoothed_per_million', 'CFR', 'CFR_per_million']], cmap = 'Blues', annot = True)

# Function to plot scatter plot
def scatter_plot(dep, dataframe):
    plt.figure()
    plt.rcParams.update({'font.size': 22})
    ind = ['total_vaccinations', 'people_vaccinated', 'people_fully_vaccinated', 'new_vaccinations', 'new_vaccinations_smoothed', 'total_vaccinations_per_hundred', 'people_vaccinated_per_hundred', 'people_fully_vaccinated_per_hundred','new_vaccinations_smoothed_per_million']
    for i, j in enumerate(ind):
        correlation = dataframe[j].corr(dataframe[dep])
        ax = dataframe.plot.scatter(x=j, y=dep, c='DarkBlue', title="x: {x_name}; y: {y_name}; corr: {corr}".format(x_name = j, y_name = dep, corr = correlation), figsize=(30,45), fontsize=22)

1. Graphic Visualization For All Countries

# Plotting heatmap 
heat_map(df)

# Plotting scatter plots: vaccination variables against total_cases
scatter_plot('total_cases', df)

<Figure size 432x288 with 0 Axes>

# Plotting scatter plots: vaccination variables against total_cases_per_million
scatter_plot('total_cases_per_million', df)

<Figure size 432x288 with 0 Axes>

# Plotting scatter plots: vaccination variables against impact
# impact = total_cases * 0.1 + hosp_patients * 0.2 + icu_patients * 0.3 + total_deaths * 0.4
scatter_plot('impact', df)

<Figure size 432x288 with 0 Axes>

# Plotting scatter plots: vaccination variables against impact_per_million
# impact_per_million = total_cases_per_million * 0.1 + hosp_patients_per_million * 0.2 + icu_patients_per_million * 0.3 + total_deaths_per_million * 0.4
scatter_plot('impact_per_million', df)

<Figure size 432x288 with 0 Axes>

2. Graphic Visualization For Malaysia

# Defining a new dataframe for Malaysia
df_my = df[df['location'] == 'Malaysia']

df_my['CFR'] = (df_my['total_cases'] / df_my['total_deaths'])*100
df_my['CFR_per_million'] = (df_my['total_cases_per_million'] / df_my['total_deaths_per_million'])*100
df_my['impact'] = (df_my['total_cases'] * 0.1 + df_my['hosp_patients'] * 0.2 + df_my['icu_patients'] * 0.3 + df_my['total_deaths'] * 0.4)
df_my['impact_per_million'] = (df_my['total_cases_per_million'] * 0.1 + df_my['hosp_patients_per_million'] * 0.2 + df_my['icu_patients_per_million'] * 0.3 + df_my['total_deaths_per_million'] * 0.4)

<ipython-input-16-995a00b07d45>:4: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  df_my['CFR'] = (df_my['total_cases'] / df_my['total_deaths'])*100
<ipython-input-16-995a00b07d45>:5: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  df_my['CFR_per_million'] = (df_my['total_cases_per_million'] / df_my['total_deaths_per_million'])*100
<ipython-input-16-995a00b07d45>:6: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  df_my['impact'] = (df_my['total_cases'] * 0.1 + df_my['hosp_patients'] * 0.2 + df_my['icu_patients'] * 0.3 + df_my['total_deaths'] * 0.4)
<ipython-input-16-995a00b07d45>:7: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  df_my['impact_per_million'] = (df_my['total_cases_per_million'] * 0.1 + df_my['hosp_patients_per_million'] * 0.2 + df_my['icu_patients_per_million'] * 0.3 + df_my['total_deaths_per_million'] * 0.4)

# Plotting heatmap 
heat_map(df_my)

# Plotting scatter plots: vaccination variables against total_cases
scatter_plot('total_cases', df_my)

<Figure size 432x288 with 0 Axes>

# Plotting scatter plots: vaccination variables against total_cases_per_million
scatter_plot('total_cases_per_million', df_my)

<Figure size 432x288 with 0 Axes>

Note: Malaysia does not provide any data on the number of hospital patients and the number of icu patients. Hence, the "impact" feature cannot be used in this context. The impact graphs below are all empty

# Plotting scatter plots: vaccination variables against impact
scatter_plot('impact', df_my)

<Figure size 432x288 with 0 Axes>

# Plotting scatter plots: vaccination variables against impact_per_million
# impact_per_million = total_cases_per_million * 0.1 + hosp_patients_per_million * 0.2 + icu_patients_per_million * 0.3 + total_deaths_per_million * 0.4
scatter_plot('impact_per_million', df_my)

<Figure size 432x288 with 0 Axes>

3. Graphic Visualization For USA

# Defining a new dataframe for USA
df_us = df[df['iso_code'] == 'USA']

df_us['CFR'] = (df_us['total_cases'] / df_us['total_deaths'])*100
df_us['CFR_per_million'] = (df_us['total_cases_per_million'] / df_us['total_deaths_per_million'])*100
df_us['impact'] = (df_us['total_cases'] * 0.1 + df_us['hosp_patients'] * 0.2 + df_us['icu_patients'] * 0.3 + df_us['total_deaths'] * 0.4)
df_us['impact_per_million'] = (df_us['total_cases_per_million'] * 0.1 + df_us['hosp_patients_per_million'] * 0.2 + df_us['icu_patients_per_million'] * 0.3 + df_us['total_deaths_per_million'] * 0.4)

<ipython-input-22-1a5ceb3b2ae5>:4: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  df_us['CFR'] = (df_us['total_cases'] / df_us['total_deaths'])*100
<ipython-input-22-1a5ceb3b2ae5>:5: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  df_us['CFR_per_million'] = (df_us['total_cases_per_million'] / df_us['total_deaths_per_million'])*100
<ipython-input-22-1a5ceb3b2ae5>:6: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  df_us['impact'] = (df_us['total_cases'] * 0.1 + df_us['hosp_patients'] * 0.2 + df_us['icu_patients'] * 0.3 + df_us['total_deaths'] * 0.4)
<ipython-input-22-1a5ceb3b2ae5>:7: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  df_us['impact_per_million'] = (df_us['total_cases_per_million'] * 0.1 + df_us['hosp_patients_per_million'] * 0.2 + df_us['icu_patients_per_million'] * 0.3 + df_us['total_deaths_per_million'] * 0.4)

# Plotting heatmap 
heat_map(df_us)

# Plotting scatter plots: vaccination variables against total_cases
scatter_plot('total_cases', df_us)

<Figure size 432x288 with 0 Axes>

# Plotting scatter plots: vaccination variables against total_cases_per_million
scatter_plot('total_cases_per_million', df_us)

<Figure size 432x288 with 0 Axes>

# Plotting scatter plots: vaccination variables against impact
# impact = total_cases * 0.1 + hosp_patients * 0.2 + icu_patients * 0.3 + total_deaths * 0.4
scatter_plot('impact', df_us)

<Figure size 432x288 with 0 Axes>

# Plotting scatter plots: vaccination variables against impact_per_million
# impact_per_million = total_cases_per_million * 0.1 + hosp_patients_per_million * 0.2 + icu_patients_per_million * 0.3 + total_deaths_per_million * 0.4
scatter_plot('impact_per_million', df_us)

<Figure size 432x288 with 0 Axes>

3.1 Testing using "new" variables

# Define new_impact feature
# new_impact = new_cases * 0.1 + hosp_patients * 0.2 + icu_patients * 0.3 + new_deaths * 0.4
df_us['new_impact'] = df_us['new_cases'] * 0.1 + df_us['hosp_patients'] * 0.2 + df_us['icu_patients'] * 0.3 + df_us['new_deaths'] * 0.4

<ipython-input-28-bfd8534c9706>:3: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  df_us['new_impact'] = df_us['new_cases'] * 0.1 + df_us['hosp_patients'] * 0.2 + df_us['icu_patients'] * 0.3 + df_us['new_deaths'] * 0.4

# Plotting scatter plots: vaccination variables against new_cases
scatter_plot('new_cases', df_us)

<Figure size 432x288 with 0 Axes>

# Plotting scatter plots: vaccination variables against hosp_patients
scatter_plot('hosp_patients', df_us)

<Figure size 432x288 with 0 Axes>

# Plotting scatter plots: vaccination variables against icu_patients
scatter_plot('icu_patients', df_us)

<Figure size 432x288 with 0 Axes>

# Plotting scatter plots: vaccination variables against icu_patients
scatter_plot('icu_patients', df_us)

<Figure size 432x288 with 0 Axes>

# Plotting scatter plots: vaccination variables against new_deaths
scatter_plot('new_impact', new_deaths)

<Figure size 432x288 with 0 Axes>